Predictive Analytics Times
Predictive Analytics Times

1 year ago
Wise Practitioner – Predictive Analytics Interview Series: Dongyang Fu and Wen Shi at Concord Advice


In anticipation of their upcoming conference presentation, Improving Credit Scoring with Hierarchical Bayesian Modeling at Predictive Analytics World for Financial in New York, Oct 29-Nov 2, 2017, we asked Dongyang Fu and Wen Shi, Data Scientists at Concord Advice, a few questions about their work in predictive analytics.

Q: In your work with predictive analytics, what behavior or outcome do your models predict?

A: There are two major fields we apply predictive modelling technique:

  1. Risk management where we try to differentiate good and bad customers based on their credit history data.
  2. Online marketing we identify higher potential customer. 

Q: How does predictive analytics deliver value at your organization – what is one specific way in which it actively drives decisions or operations?

A: We are able to optimize marketing and operating resources based on our modelling results and aim to improve performance on both revenue and cost sides.

Q: Can you describe a quantitative result, such as the predictive lift of your model or the ROI of an analytics initiative?

A: We do not disclose quantitative numbers, but we have seen huge drawdown after deploying bad models due to data integrity and how much improvement were made after we corrected the model. The conclusion is that a healthy functioning model is crucial to the company’s success.

Q: What surprising discovery or insight have you unearthed in your data?

A: Superiority of a specific methodology exists when the data structure and the information within better fits the methodology’s assumptions. Performance degradation can be somehow overcome by score fusion of different types of models.

Q: Sneak preview: Please tell us a take-away that you will provide during your talk at Predictive Analytics World.

A: We have been using traditional predictive modelling techniques forever, logistic modeling techniques has been used for a long time. Until recently we started exploring other methods that could help us get more insight about the data and the customers, the team has identified some alternative approaches which require different type of data input. With more data products available, we need to be adapting to the changes. Our talk not only focus on how new techniques improve business performances, but more about how different techniques give us a new dimension in terms of understanding underlying behavior of the people.


Don’t miss Dongyang and Wen’s conference presentation, Improving Credit Scoring with Hierarchical Bayesian Modeling on Monday, October 30, 2017 from 3:05 to 3:25 pm at Predictive Analytics World for Financial in New York. Click here to register to attend. Use Code PATIMES for 15% off current prices (excludes workshops).

By: Eric Siegel, Founder, Predictive Analytics World

Eric Siegel is the founder of Predictive Analytics World ( — the leading cross-vendor conference series consisting of 10 annual events in New York, Chicago, San Francisco, Washington DC, London, and Berlin — and the author of the award-winning book, Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie, or Die – Revised and Updated Edition, (Wiley, 2016).

Leave a Reply